Hybrid FRP-Concrete-Steel Double-Skin Tubular Structural Members

Transcription

1 CICE 21 - The 5th International Conference on FRP Composites in Civil Engineering September 27-29, 21 Beijing, China Hybrid FRP--Steel Double-Skin Tubular Structural Members J.G. Teng (cejgteng@inet.polyu.edu.hk), T. Yu and Y.L. Wong Department of Civil and Structural Engineering, The Hong Kong Polytechnic University, Hong Kong, China ABSTRACT: Hybrid FRP-concrete-steel double-skin tubular members (DSTMs) are a new form of hybrid structural members composed of an inner steel tube, an outer FRP tube and a concrete infill between them. The two tubes may be concentrically placed to produce a section form more suitable for columns, or eccentrically placed to produce a section form more suitable for beams. These hybrid structural members possess many important advantages over conventional structural members, including their excellent corrosion resistance as well as excellent ductility and seismic resistance. This paper explains the rationale and advantages of this new form of structural members and discusses their potential applications before presenting a summary of the recent and current studies on their structural behaviour and design. These studies form part of a major on-going research programme at The Hong Kong Polytechnic University (PolyU). 1 INTRODUCTION Hybrid FRP-concrete-steel double-skin tubular members (referred to as hybrid DSTMs) (Figure 1) are a new form of hybrid members proposed by the first author (Teng et al. 24, 27). A hybrid DSTM consists of an outer tube made of fiber-reinforced polymer (FRP) and an inner tube made of steel, with the space between filled with concrete. The two tubes may be concentrically placed (Figures 1a and 1b) to produce a section form more suitable for columns, or eccentrically placed to produce a section form more suitable for beams (Figures 1c and 1d). In hybrid DSTMs, the FRP tube offers mechanical resistance primarily in the hoop direction to confine the concrete and to enhance the shear resistance of the member. Hybrid DSTMs may be constructed insitu or precast, with the two tubes acting as the stayin-place form. The sections of the two tubes may be both circular (Figures 1a and 1c), rectangular (Figure 1d), or in another shape; they may also have shapes different from each other (Figure 1b). Shear connectors need to be provided between the steel tube and the concrete, particularly in beams, but are generally not needed for the FRP tube which is normally designed to have only a small longitudinal stiffness. The most important advantage of hybrid DSTMs is their excellent corrosion resistance, as the FRP tube is highly resistant to corrosion while the steel tube is protected by the FRP tube and the concrete. The other main advantages of hybrid DSTMs include (1) excellent ductility, as the concrete is well confined by the two tubes and inward local buckling of the steel tube is constrained by the concrete; (2) a high strength/stiffness-to-weight ratio as the inner void largely eliminates the redundant concrete; (3) ease for construction, as the two tubes act as a permanent form for casting concrete, and the presence of the inner steel tube and the concrete allows easy connection to other members. More discussions of the rationale and advantages of hybrid DSTMs are available in Teng et al. (27). (a) FRP tube Steel tube Con- FRP tube Steel tube (c) Figure 1. Typical sections of hybrid DSTMs (b) (d)

2 2 POTENTIAL PRACTICAL APPLICATIONS FRP deck 2.1 Double-skin tubular columns Because of their excellent corrosion resistance, hybrid DSTMs are most suitable for use in outdoor structures such as bridges and coastal structures which are likely to be exposed to a harsh environment. Hybrid DSTMs can be used as compression members, such as piles and bridge piers, various towers (e.g. wind turbine towers, electricity transmission towers and bridge towers) and other similar structures. Hereafter hybrid DSTMs used as compression members are referred to as hybrid doubleskin tubular columns or hybrid DSTCs for brevity. In practical applications, when the length of DSTCs becomes very large (e.g. as wind turbine towers), they can be constructed using a segmental method, which involves the segmental construction of the outer FRP tube and the inner steel tube and the use of the two tubes as the permanent formwork for the in-situ casting of concrete as schematically illustrated in Figure 2. The flanges of the steel tubular segments and the FRP profiles inside the FRP tubular segments serve not only as longitudinal connectors between the segments, but also as (1) shear connectors between the concrete and the tubes; and (2) stiffeners to both tubes, thus improving the structural performance of hybrid DSTCs. Steel tube FRP tube Adhesive Layer (a) FRP bars Beam/deck connectors Shear studs Pre-embedded reinforcement Steel tube (b) FRP tube Figure 3. Hybrid DSTB/deck units Figure 2. Segmental construction of tall DSTCs 2.2 Double-skin tubular beams Hybrid DSTMs can also be used as flexural members in bridges, costal structures and other structures exposed to a harsh environment. Hereafter hybrid DSTMs used as flexural members are referred to as hybrid double-skin tubular beams or hybrid DSTBs for brevity. As bridge girders, hybrid DSTBs can be used with an all FRP deck (or another form of lightweight and corrosion-resistant deck such as a hybrid FRPconcrete deck or an aluminium deck) to form a light slab-on-girder bridge system (Figure 3a). In Figure 3a, the deck is connected with the hybrid DSTB through an adhesive layer, but this adhesive layer can be replaced by or used together with mechanical joints (e.g. bonded shear keys) when necessary. Hybrid DSTBs can also be integrated into a concrete deck reinforced with FRP bars to form a bridge system (Figure 3b). In such cases, pre-embedded steel (stainless steel may be used here) reinforcement in the DSTB can be spliced with the bottom layer of FRP bars in the deck using mechanical couplers for beam-deck connection (Figure 3b). Additional shear connectors in the form of U-shaped dowels (or other appropriate forms) passing through the FRP tube may also be used to ensure the longitudinal compos-

3 ite action between the beam and the deck (Figure 3b). As the fibres in the FRP tube are close to the hoop direction, the passing of steel bars through the tube is expected to affect little its overall performance. While hybrid DSTBs with a constant crosssectional configuration are particularly suitable for simply-supported beams, cost-effective continuous DSTBs can be built by placing the steel inner tube in an inclined position in the longitudinal direction (i.e. the steel tube lies near the bottom of the crosssection in sagging moment zones but near the top of the cross-section in hogging moment zones). For a beam/deck unit as illustrated in Figure 3a, if composite action between the DSTB and the deck is not pursued (e.g. because of the relatively low stiffness/strength in the longitudinal direction of an all FRP deck), the behaviour of the DSTB alone governs the performance in the longitudinal direction of the beam/deck assembly. For a beam/deck unit as illustrated in Figure 3b, the composite action between the beam and the deck typically needs to be considered. 3 EXISTING RESEARCH ON HYBRID DSTMS At The Hong Kong Polytechnic University (PolyU), a large amount of research has been conducted on hybrid DSTMs. Most of the work undertaken prior to 27 can be found in Teng et al. (27), Yu et al. (26), Wong et al. (28), and Yu et al. (21a, 21b, 21c, 21d). Teng et al. (27) explained in detail the rationale for the new member form together with its expected advantages, and presented preliminary experimental results to demonstrate some of the advantages of this new member form, such as excellent ductility and shear resistance. Yu et al. (26) presented the results of a systematic experimental study on the flexural behavior of hybrid DSTMs as well as results from a corresponding theoretical model based on the fiber element approach. Yu et al. (26) showed that the flexural response of hybrid DSTMs, including their flexural stiffness, cracking load and ultimate load, can be substantially improved by shifting the inner steel tube towards the tension zone or by providing FRP bars as additional longitudinal reinforcement. Wong et al. (28) presented a systematic experimental study on the compressive behavior of hybrid DSTMs and compared the performance of hybrid DSTMs with that of FRP-confined solid cylinder/column (FCSC) specimens and FRP-confined hollow cylinder/column (FCHC) specimens; a good understanding of the behavior of concrete in hybrid DSTMs resulted from this study. Yu et al. (21a, b) developed a new plastic-damage model for FRPconfined concrete based on a critical review of the previous D-P type plasticity models. A finite element model incorporating the new plastic-damage model was shown to provide close predictions of the test results of hybrid DSTMs (Yu et al. 21b). Based on the available experimental observations and the results from the finite element model, Yu et al. (21c) proposed a design-oriented stress-strain model for the confined concrete in hybrid DSTMs subjected to axial compression. Yu et al. (21d) presented experimental results on the behaviour of hybrid DSTMs subjected to eccentric compression as well as a so-called variable confinement model for the confined concrete to account for the effect of strain gradient on confinement effectiveness. These studies at PolyU have led to a simple design approach for hybrid DSTMs as columns, and this design approach has recently been adopted by the draft Chinese Code for Infrastructure Application of FRP Composites (Teng and Yu 21). In addition to research carried out at PolyU, work has also been undertaken by Xu and Tao (25), Yu (26), Liu (27) and Han et al. (21). These studies have further confirmed the good performance of hybrid DSTMs subjected to different loading conditions. At PolyU, intensive research on the behaviour and design of hybrid DSTMs is continuing. In the remainder of the paper, the latest advances of this ongoing research programme at PolyU are briefly presented, covering: (1) cyclic axial compression tests; (2) hybrid DSTCs with high strength concrete; (3) large-scale column tests; (4) behaviour of hybrid DSTBs. 4 CYCLIC AXIAL COMPRESSION TESTS Since hybrid DSTCs are particularly suitable for use in seismic regions, their behaviour under cyclic loading is of significant interest. Therefore, a series of cyclic axial compression tests on hybrid DSTCs has recently been completed as part of a comprehensive study aiming at the eventual development of an appropriate procedure for the seismic design of hybrid DSTCs. In total, eight identical hybrid DSTCs were tested, covering four loading schemes; two specimens were made for each loading scheme. The specimens had an outer diameter (i.e. the outer diameter of the annular concrete section) of 25.3 mm, an inner diameter (i.e. the inner diameter of the annular concrete section and the outer diameter of the inner steel tube) of 14.3 mm, and a height of 4 mm. The void ratio of these specimens (i.e. ratio between the inner diameter and the outer diameter of the annular concrete section) was thus around.7. The nominal thickness of the FRP tube formed from fibre sheets via the wet layup process was.34 mm while the thickness of the steel tube was 5.3 mm. The use of wet layup FRP tubes with fibres oriented solely in

4 Axial Stress (MPa) Axial Load (kn) the hoop direction instead of filament-wound FRP tubes with fibres oriented at small angles to the hoop direction was due to difficulties in finding commercial FRP tubes with properties required by the tests. Such use is considered to be acceptable as the main function of the FRP tube is to provide hoop resistance. The four loading schemes included monotonic loading, full unloading/reloading, partial unloading and partial reloading. More details of the experimental programme are available in Yu et al. (21e) M1 5 M2 F1 F Axial Strain Figure 4. Axial load-axial strain curves of DSTCs under cyclic axial compression The axial load-axial strain curves of the two specimens subjected to full unloading/reloading cycles (i.e. specimens F1 and F2) are compared with those of the two monotonic loading specimens (i.e. specimens M1 and M2) in Figure 4. It is evident from Figure 4 that the envelope curves of the two specimens subjected to cyclic axial compression, which provide an upper boundary of their responses under cyclic loading, are almost the same as the axial load-axial strain curves of the two specimens subjected to monotonic compression. Figure 4 also shows that the unloading/reloading cycles at the same unloading displacement generally do not coincide with each other, indicating that the effect of loading history on the cyclic response of hybrid DSTCs is not negligible. The difference between two consecutive loading cycles is, however, seen to become smaller as the number of repeated cycles increases (Figure 4). The axial stress-axial strain curves of the confined concrete in specimens M1, M2, F1 and F2 are shown in Figure 5. The axial stress of concrete in a DSTC is defined herein as the load carried by the annular concrete section divided by its crosssectional area. The load carried by the concrete section is assumed to be equal to the difference between the load carried by the DSTC and that carried by the steel tube; the latter was found from compression tests on hollow steel tubes. These tests on hybrid DSTCs subjected to cyclic axial compression revealed that during unloading, the steel tube unloaded more rapidly than the concrete and reached the zero stress first when considerable compressive stresses still existed in the concrete. As unloading continued, tensile stresses were induced in the steel tube. In deducing the axial stress in the concrete, it was assumed that the ratio of axial stress increment to axial strain increment for the steel is equal to the unloading modulus of steel found from the cyclic compression test of a hollow steel tube cut from the same original long tube which provided steel tubes for the hybrid DSTCs tested. The above difference in unloading behaviour between the steel tube and the concrete is at least partially due to their different modulus degradations: the unloading modulus of a steel tube is generally the same as its elastic modulus but the unloading modulus of concrete is considerably smaller than its elastic modulus due to degradation M1 M2 F1 F Axial Strain Figure 5. Axial stress-axial strain curves of concrete in DSTCs under cyclic axial compression Figure 5 shows that the envelope curves of specimens F1 and F2 are almost the same as the axial stress-axial strain curve of specimens M1 and M2. It is also interesting to note that while specimens F1 and F2 were subjected to full unloading/reloading cycles (i.e. unloading was terminated at a very small load), the axial stresses at the termination points of unloading were significantly larger than zero. This phenomenon is due to the development of tensile stresses in the steel tube, as explained earlier. More details of the test results and discussions of the effect of partial unloading and partial reloading are available in Yu et al. (21e). 5 HYBRID DSTCS WITH HIGH STRENGTH CONCRETE High strength concrete (HSC) is normally defined as concrete with a cylinder compressive strength exceeding 5 MPa (Xiao et al. 21). Hybrid DSTCs offer an ideal opportunity for the use of HSC: the

5 Axial load ( kn) confinement provided by the two tubes ensures a ductile response of the concrete and the absence of any steel bars ensures good-quality casting of HSC. Six hybrid DSTC specimens with HSC have recently been tested under axial compression at PolyU, as part of a comprehensive study. The concrete of these specimens had a cylinder compressive strength of 83.5 MPa. These six specimens covered three combinations of void ratio and FRP tube; two identical specimens were prepared for each combination. Two void ratios (i.e..558 and.779) within the practical range were adopted. Two types of FRP tubes, namely, one-ply carbon FRP (CFRP) tubes and six-ply glass FRP (GFRP) tubes, were used; the two types of FRP tubes had a similar circumferential stiffness but the GFRP tubes had a much larger ultimate hoop strain (around two times that of the CFRP tubes). All specimens had the same outer diameter of 24 mm and the same height of 4 mm. 3 The axial load-axial strain curves of all six specimens are shown in Figure 6. While all these curves are close to an approximately elasticperfectly-plastic curve, the plastic plateau for those specimens with a GFRP tube is much longer than that for specimens with a CFRP tube. The two specimens with a GFRP tube showed a very ductile response with the ultimate axial strain being over 1.6% and around 5 times the strain of unconfined concrete at peak stress (i.e..329%, found from tests on three mm x 35 mm concrete cylinders). By contrast, the ultimate axial strain is only around.7% for the four specimens with a CFRP tube. The two different void ratios (i.e..558 and.779) did not lead to obvious differences in the ultimate axial strain of hybrid DSTCs with a CFRP tube. These test results suggest that hybrid DSTCs with HSC can have very good ductility when the FRP outer tube has a sufficiently large strain capacity in the circumferential direction. This advantage, however, may be much less pronounced when a relatively brittle FRP tube is used. Further work on hybrid DSTCs with HSC is being conducted at PolyU, covering a larger range of concrete strength and aiming to develop a unified stress-strain model for both NSC and HSC in hybrid DSTCs. 6 LARGE-SCALE COLUMN TESTS The early experimental studies undertaken at PolyU and elsewhere were limited to the testing of smallscale columns. Whether the conclusions reached on the basis of these small-scale tests are valid for fullscale members is an issue of concern. To address this concern, an experimental study is under way where large-scale columns are tested under pure axial compression or combined axial compression and lateral loading. Preliminary test results of three large-scale hybrid DSTCs under pure axial compression are presented below Void Ratio=.779, CFRP Tube 1 Void Ratio=.779, CFRP Tube Void Ratio=.779, GFRP Tube 5 Void Ratio=.779, GFRP Tube Void Ratio=.588, CFRP Tube Void Ratio=.588, CFRP Tube Axial strain Figure 6. Axial load-axial strain curves of hybrid DSTCs with high strength concrete Figure 7. Large-scale hybrid DSTC after test The three specimens all had an outer diameter of 44 mm and a height of 8 mm. Two of the specimens (i.e. Specimens S325-4 and S325-6) had the same steel tube with an outer diameter of 324 mm and a thickness of 9 mm, leading to a void ratio of.8; the two specimens had an FRP tube with a nominal thickness of.68 mm (i.e. a 4-ply FRP tube) and 1.2 mm (i.e. a 6-ply FRP tube) respectively. The remaining specimen had a steel tube with an outer diameter of 246 mm and a thickness of 8 mm, a void ratio of.61, and a 6-ply FRP tube (i.e. Specimen S245-6). Once again, the FRP tubes were formed from dry fibre sheets in the laboratory using the wet layup method with all fibres oriented in the hoop direction. Similar to the small-scale specimens tested by Wong et al. (28), all three large-scale specimens failed by the rupture of the FRP tube as a result of hoop tension. One of the large-scale specimens after test is shown in Figure 7.

6 Axial load (kn) The axial load-axial strain curves of these three DSTC specimens are shown in Figure 8, where the strains were found from the readings of displacement transducers covering the mid-height region of 32 mm. All three specimens showed an approximately bilinear axial load-axial strain curve. The slope of the second linear portion of the curve is seen to increase significantly with the thickness of the FRP tube, but the void ratio does not seem to affect this slope significantly. For the two DSTCs with a six-ply FRP tube, the ultimate axial strains are over ten times the strain at the peak stress of unconfined concrete. The much smaller ultimate axial strain for the specimen with a four-ply FRP tube was partially caused by the unexpected early failure of the FRP tube at a relatively small hoop strain due to a defect arising from the preparation of the specimen. These preliminary test results suggest that the behaviour of large-scale hybrid DSTCs can be extrapolated from that found from testing small-scale hybrid DSTCs (Wong et al. 28). While further studies are needed to examine the effect of member size in various terms such as the axial stress-axial strain curve of the confined concrete and the ultimate state of the FRP tube, these preliminary test results have confirmed the good performance of hybrid DSTCs as initially expected S245-6 S325-6 S Axial strain Figure 8. Axial load-axial strain curves of large-scale DSTCs 7 BEHAVIOUR OF HYBRID DSTBS As mentioned earlier, existing studies at PolyU and elsewhere have been focused on the behaviour of hybrid DSTMs as columns (i.e. hybrid DSTCs). Only a limited number of laboratory tests have been conducted on small-scale hybrid DSTBs without shear connectors and subjected to static loads (Yu et al. 26; Liu 27). Among these tests, ten specimens had a section form as shown in Figure 1a (i.e. two concentric circular tubes) and were tested as part of a study to understand the beam-column behaviour of these hybrid members; the other five had a section form as shown in Figure 1c which is more suitable for use in beams. Further research is being pursued to address the issues outlined below. Yu et al. (26) showed that shear connectors should be provided between the steel tube and the concrete in hybrid DSTBs to ensure a high degree of composite action between them. Research is therefore needed to establish appropriate forms of shear connectors and to develop a design method for such shear connectors. In hybrid DSTBs, the compressive concrete is subjected to lateral conferment by the FRP tube and the steel tube, so the behaviour of the shear connectors in hybrid DSTBs is expected to be different from that of shear connectors in traditional hybrid steel-concrete beams with a steel I-section. Existing design methods developed for the latter (e.g. Eurocode 4 24) are thus not expected to be applicable to the former. To address this issue, appropriately-designed push-out tests need to be conducted to investigate the behaviour of shear connectors and to develop simple design methods for them. For DSTBs to be used as bridge girders, research on their fatigue behaviour is needed. As fatigue failure of a hybrid beam generally occurs in the shear connectors, the behaviour of these connectors under fatigue loading needs to be understood first through push-out tests. A reliable analysis approach is also needed to predict the fatigue life of hybrid DSTBs, which should appropriately account for partial interaction (Newmark et al. 1951), fatigue-induced material property changes (e.g. softening and/or creep deformation of concrete) (Holmen 1982; Ahmad et al. 24), and fatigue-induced stiffness reduction of shear connectors. Although hybrid DSTBs are expected to be very strong in shear because of the presence of an inner steel tube and an outer FRP tube, their high shear strength needs to be confirmed through laboratory testing. A simple design method also needs to be developed for the shear strength of hybrid DSTBs, where the simple additive approach (Teng et al. 22; Chen and Teng 23) widely used for FRPstrengthened RC beams may be adopted. 8 CONCLUDING REMARKS This paper has discussed the rationale and advantages of hybrid FRP-concrete-steel double-skin tubular members (i.e. hybrid DSTMs), provided a brief summary of existing research on hybrid DSTMs, and outlined the issues currently being addressed by the authors research group in a major on-going research programme at The Hong Kong Polytechnic University. Hybrid DSTMs have a great potential for use as piles, columns and towers as well as beams in structures exposed to harsh environments such as bridges, costal structures and various tower struc-

7 tures. While existing and current research is mainly concerned with DSTMs as isolated members, exciting opportunities exist for the development of structural systems based on DSTMs. One such example is the development of a bridge system with DSTCs as piers, piles and towers and with DSTBs as girders supporting corrosion-resistant decks. Such a bridge system is highly durable and offers an attractive alternative to existing structural systems in a corrosive environment. Such system-level research is needed in the near future to realize the full potential of DSTMs. 9 ACKNOWLEDGMENTS The authors are grateful for the financial support received from the Research Grants Council of the Hong Kong SAR (Project No: PolyU 5278/7E) and The Hong Kong Polytechnic University. They are also grateful to Mr Y.B. Cao, Ms P. Xie and Mr B. Zhang who conducted the tests described in Sections 4 to 6. 1 REFERENCES Ahmad, I., Zhu, Z. and Mirmiran, A. 28. Fatigue behaviour of concrete-filled fiber-reinforced polymer tubes, Journal of Composites for Construction, ASCE, 12(4): Chen, J.G. and Teng, J.G. 23. Shear capacity of fiberreinforced polymer-strengthened reinforced concrete beams: Fiber reinforced polymer rupture, Journal of Structural Engineering, ASCE, 129(5): Eurocode 4 24, Design of Composite Steel and Structures, European Committee for Standardization (CEN). Han, L.H., Tao, Z., Liao, F.Y. and Xu, Y. 21. Tests on cyclic performance of FRP-concrete-steel double-skin tubular columns, Thin-Walled Structures, 48(6): Holmen, J Fatigue of concrete by constant and variable amplitude loading, Fatigue of Structures, ACI SP-75, Detriot, MI, Liu, M.X. 27. Studies on Key Behavior of FRP-- Steel Double Skin Tubular Columns. PhD Thesis, Tsinghua University. (in Chinese). Newmark, N.M., Siess, C.P. and Viest, I.M Tests and analysis of composite beams with incomplete interaction, Proceedings, Society for Experimental Stress Analysis, 9(1): Teng, J.G. and Yu, T. 21. Hybrid FRP-concrete tubular columns: design provisions in the draft Chinese code, Proceedings, International Symposium on Life-Cycle Performance of Bridges and Structures, Changsha, China, June, pp Teng, J.G., Chen, J.F., Smith, S.T. and Lam, L. 22. FRP- Strengthened RC Structures, John Wiley and Sons, UK. Teng, J.G., Yu, T. and Wong, Y.L. 24. Behaviour of hybrid FRP-concrete-steel double-skin tubular columns, Proceedings, Second International Conference on FRP Composites in Civil Engineering, December 8-1, Adelaide, Australia, Teng, J.G., Yu, T., Wong, Y.L. and Dong, S.L. 27. Hybrid FRP-concrete-steel tubular columns: concept and behavior, Construction and Building Materials, 21(4): Wong, Y.L., Yu, T., Teng, J.G. and Dong, S.L. 28. Behavior of FRP-confined concrete in annular section columns, Composites: Part B-Engineering, 39: Xiao, Q.G., Teng, J.G. and Yu, T. (21). Behavior and modeling of confined high strength concrete, Journal of Composites for Construction, ASCE, 14 (3): Xu, Y. and Tao, Z. 25. Key issues of dynamic behavior of hybrid FRP-concrete-steel double skin tubular columns, Journal of Fuzhou University (Natural Science), 33: (in Chinese). Yu, T., Wong, Y.L., Teng, J.G., Dong, S.L. and Lam, S.S. 26. Flexural behaviour of hybrid FRP-concretesteel double skin tubular members, Journal of Composites for Construction, ASCE, 1 (5): Yu, T., Teng, J.G., Wong, Y.L. and Dong, S.L. 21a. Finite element modeling of confined concrete-i: Drucker- Prager type plasticity model, Engineering Structures, 32 (3): Yu, T., Teng, J.G., Wong, Y.L. and Dong, S.L. 21b. Finite element modeling of confined concrete-ii: plasticdamage model, Engineering Structures, 32(3): Yu, T., Teng, J.G. & Wong, Y.L. 21c. Stress-strain behavior of concrete in hybrid double-skin tubular columns, Journal of Structural Engineering, ASCE, 136(4): Yu, T., Wong, Y.L. and Teng, J.G. 21d. Behavior of hybrid FRP-concrete-steel double-skin tubular columns subjected to eccentric compression, Advances in Structural Engineering, 13(5). Yu,T., Teng, J.G., Zhang, B. and Cao, Y.B. 21e. Behavior of hybrid FRP-concrete-steel double-skin tubular columns subjected to cyclic axial compression, in preparation. Yu, X.W. 26. Behavior of Hybrid CFRP--Steel Double-Skin Tubular Columns under Axial Compression. MSc Thesis, Harbin Institute of Technology. (in Chinese).